Objective of the proposed research is to advance the field of unsupervised visual anomaly detection (UAD), with a focus on medical imagery. Developed UAD methods will be evaluated on all the available medical imagery available as part of the Pediatric Brain Tumor Atlas, where anomaly detection (e.g. tumor, metastases) is applicable (e.g. histology images, radiology images, MR images). In comparison with existing supervised/weakly-supervised approaches, unsupervised approaches will not require expertly (e.g. lesion level) labeled data and will be more widely applicable to the vast amount of the existing medical imagery data.
What are the goals of this project?
Research will be performed as part of the iPC EU Horizon 2020 project1, where one of the main objectives of the proposed research is to support individualized diagnostics and treatment.
The Children's Brain Tumor Network contributed to this project by providing access to the Pediatric Brain Tumor Atlas.
PI: Dejan Stepec
Adam Resnick, PhD
Adam Resnick is the Director of Data Driven Discovery in Biomedicine (D3b) at Children’s Hospital of Philadelphia (CHOP) responsible for leading a multidisciplinary team to build and support a scalable, patient-focused healthcare and educational discovery ecosystem on behalf of all children. He i
Children’s Hospital of Philadelphia